Immunotherapy Safety for the Primary Care ... - U.S. Coast Guard

J ALLERGY CLIN IMMUNOL
VOLUME 115, NUMBER 3
Lieberman et al S517
istry of penicillin is well characterized, validated skin
testing reagents representing the various allergenic determinants
have been developed. In large-scale studies about
90% of patients with a history of penicillin allergy have
negative penicillin skin test responses. 200,201
The positive predictive value of penicillin skin testing
is 50% or greater. 201,202 Patients with positive penicillin
skin test responses should receive an alternate antibiotic
or undergo rapid desensitization if administration of
penicillin is mandated. The negative predictive value of
penicillin skin testing (for immediate-type reactions) is
between 97% and 99%, depending on the skin test
reagents used. 200,201,203 Patients with negative penicillin
skin test responses might be safely treated with penicillin,
and depending on the reagents used for skin
testing, the therapeutic dose might be preceded by a test
dose.
Penicillin skin testing is safe in that the risk of inducing
serious reactions during properly performed penicillin
skin testing is comparable with the risk of other types of
skin testing. 204 Penicillin skin testing itself might sensitize
a very small proportion of patients. 205 Skin testing with
semisynthetic penicillins, such as ampicillin or amoxicillin,
is not standardized, and its predictive value is
unknown. Penicillin skin testing should not be performed
on patients with histories of severe non–IgE-mediated
allergic reactions to penicillin, such as Stevens-Johnson
syndrome or toxic epidermal necrolysis.
Cephalosporins. Penicillins and cephalosporins share
a common b-lactam ring, but the extent of allergic crossreactivity
between the 2 families appears to be relatively
low. Recent studies demonstrated no serious allergic
reactions in large groups of patients with a history of
penicillin allergy who were treated with cephalosporins.
206,207 Patients in these retrospective studies, however,
were given diagnoses of penicillin allergy on the basis of
self-report. Patient history is known to be poor predictor of
true penicillin allergy in that about 90% of patients with
such a history turn out to have negative penicillin skin test
responses and are able to tolerate penicillin. 200,201 A
review of the published literature showed that among
patients with a history of penicillin allergy who were
proved to have positive penicillin skin test responses, only
a small percentage of patients experienced an allergic
reaction on being challenged with cephalosporins. 208
However, fatalities have occurred when patients are not
skin tested for penicillin and given cephalosporins. 209
There are distant case reports of cephalosporin-induced
anaphylactic reactions in patients with a history of
penicillin allergy, 210,211 but these patients did not undergo
penicillin skin testing, and early cephalosporins were also
known to contain trace amounts of penicillin.
Patients with a history of penicillin allergy who have
negative penicillin skin test responses might receive
cephalosporins because they are at no higher risk of
experiencing allergic reactions. 212 In patients with a history
of penicillin allergy who have positive penicillin skin
test responses, the physician has 3 options: (1) administration
of an alternate non–b-lactam antibiotic; (2) administration
of a cephalosporin through graded challenge; or
(3) desensitization to the cephalosporin. 212
Other b-lactam antibiotics. Monobactams (aztreonam)
do not-cross react with penicillin or other b-lactams, aside
from ceftazadime, with which it shares an identical R-
group side chain. 213 Therefore patients allergic to penicillin
and other b-lactams (except for ceftazidime) might
safely receive aztreonam. Similarly, patients allergic to
aztreonam might safely receive other b-lactams, except
for ceftazidime.
Skin test studies indicate allergic cross-reactivity
between carbapenems and penicillin. 214 Although clinical
challenge studies in patients with penicillin allergy are
lacking, carbapenems should be considered cross-reactive
with penicillin.
Non–b-lactam antibiotics. Non–b-lactam antibiotics
appear to be uncommon causes of anaphylactic reactions.
Diagnosis of IgE-mediated allergy to these drugs is more
difficult because of lack of knowledge (in most cases) of
the relevant metabolites and allergenic determinants. Skin
testing with the native antibiotic can yield some useful
information because if a nonirritating concentration is
used, a positive result suggests the presence of drugspecific
IgE antibodies. 215 However, the positive predictive
value of such testing is unknown, and the negative
predictive value is even less certain. Therefore diagnosis
of anaphylactic reactions to non–b-lactam antibiotics is
primarily based on the patient’s history.
Aspirin and nonsteroidal anti-inflammatory drugs.
Aspirin and nonsteroidal anti-inflammatory drugs
(NSAIDs), including COX-2–specific inhibitors, have
all been described to cause anaphylactic reactions.
Aspirin and NSAIDs appear to be the second most
common cause of drug-induced anaphylaxis (after penicillin).
2,216 Anaphylactic reactions are unrelated to other
reactions caused by these drugs, such as respiratory
reactions and exacerbations of chronic idiopathic urticaria.
217 Although the reactions are referred to as anaphylactic,
in most cases efforts to detect drug-specific IgE
antibodies (through skin testing or in vitro testing) have
been unsuccessful. The reactions are assumed to be
anaphylactic because generally patients are able to tolerate
the drug for a period of time before a reaction ensues.
Anaphylactic reactions to aspirin and NSAIDs appear to
be medication specific in that allergic patients are able to
tolerate other NSAIDs, but this is largely based on clinical
experience rather than large-scale challenge studies. 217
Cancer chemotherapeutic agents. Anaphylaxis to
anticancer chemotherapy drugs is being encountered
more frequently because use of these drugs has increased,
218 particularly the platinum-containing drugs,
such as cisplatinum and carboplatinum. In some instances
the solvent in which these drugs are formulated
(Cremophor-L) might cause an anaphylactoid reaction. 219
Such anaphylactoid reactions to the drug product must be
distinguished from anaphylaxis because of the drug. Skin
testing to these agents is helpful in determining whether
sensitivity exists and at what dose to proceed with
sensitization if this is necessary. 220